diff --git a/brozzler/__init__.py b/brozzler/__init__.py index 7dd284d..4a1c289 100644 --- a/brozzler/__init__.py +++ b/brozzler/__init__.py @@ -2,7 +2,7 @@ brozzler/__init__.py - __init__.py for brozzler package, contains some common code -Copyright (C) 2014-2017 Internet Archive +Copyright (C) 2014-2024 Internet Archive Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. @@ -20,7 +20,7 @@ limitations under the License. import logging from pkg_resources import get_distribution as _get_distribution -__version__ = _get_distribution("brozzler").version +__version__ = _get_distribution("brozzler_ia").version class ShutdownRequested(Exception): @@ -344,6 +344,7 @@ EPOCH_UTC = datetime.datetime.utcfromtimestamp(0.0).replace(tzinfo=doublethink.U # we could make this configurable if there's a good reason MAX_PAGE_FAILURES = 3 +import brozzler_ia as brozzler from brozzler.worker import BrozzlerWorker from brozzler.robots import is_permitted_by_robots from brozzler.frontier import RethinkDbFrontier diff --git a/brozzler/browser.py b/brozzler/browser.py index 9c72589..e2914a8 100644 --- a/brozzler/browser.py +++ b/brozzler/browser.py @@ -18,13 +18,12 @@ limitations under the License. import logging import time -import brozzler +import brozzler_ia as brozzler import itertools import json import websocket import time import threading -import brozzler from requests.structures import CaseInsensitiveDict import datetime import base64 diff --git a/brozzler/chrome.py b/brozzler/chrome.py index a83632b..4dc426f 100644 --- a/brozzler/chrome.py +++ b/brozzler/chrome.py @@ -22,7 +22,7 @@ import time import threading import subprocess import os -import brozzler +import brozzler_ia as brozzler import select import re import signal diff --git a/brozzler/cli.py b/brozzler/cli.py index 3cb7c9a..3ad8f6d 100755 --- a/brozzler/cli.py +++ b/brozzler/cli.py @@ -18,7 +18,7 @@ limitations under the License. """ import argparse -import brozzler +import brozzler_ia as brozzler import brozzler.worker import datetime import json diff --git a/brozzler/dashboard/__init__.py b/brozzler/dashboard/__init__.py index 16e34c1..5729b6a 100644 --- a/brozzler/dashboard/__init__.py +++ b/brozzler/dashboard/__init__.py @@ -325,6 +325,7 @@ except ImportError: def main(argv=None): import argparse + import brozzler_ia as brozzler import brozzler.cli argv = argv or sys.argv diff --git a/brozzler/easy.py b/brozzler/easy.py index f8392f6..f846505 100644 --- a/brozzler/easy.py +++ b/brozzler/easy.py @@ -25,6 +25,7 @@ try: import warcprox import warcprox.main import pywb + import brozzler_ia as brozzler import brozzler.pywb import wsgiref.simple_server import wsgiref.handlers diff --git a/brozzler/frontier.py b/brozzler/frontier.py index afb2a57..751f148 100644 --- a/brozzler/frontier.py +++ b/brozzler/frontier.py @@ -17,7 +17,7 @@ limitations under the License. """ import logging -import brozzler +import brozzler_ia as brozzler import random import time import datetime diff --git a/brozzler/model.py b/brozzler/model.py index fe9f8c0..2596c0f 100644 --- a/brozzler/model.py +++ b/brozzler/model.py @@ -17,7 +17,7 @@ See the License for the specific language governing permissions and limitations under the License. """ -import brozzler +import brozzler_ia as brozzler import base64 import cerberus import copy diff --git a/brozzler/pywb.py b/brozzler/pywb.py index 8b61841..106c1de 100644 --- a/brozzler/pywb.py +++ b/brozzler/pywb.py @@ -41,7 +41,7 @@ import doublethink import rethinkdb as rdb import urlcanon import json -import brozzler +import brozzler_ia as brozzler import argparse r = rdb.RethinkDB() diff --git a/brozzler/robots.py b/brozzler/robots.py index 744c996..acdb3f2 100644 --- a/brozzler/robots.py +++ b/brozzler/robots.py @@ -24,7 +24,7 @@ limitations under the License. import json import logging -import brozzler +import brozzler_ia as brozzler import reppy import reppy.cache import reppy.parser diff --git a/brozzler/worker.py b/brozzler/worker.py index 116bb75..d6498ea 100644 --- a/brozzler/worker.py +++ b/brozzler/worker.py @@ -19,7 +19,7 @@ limitations under the License. """ import logging -import brozzler +import brozzler_ia as brozzler import brozzler.browser import threading import time diff --git a/brozzler/ydl.py b/brozzler/ydl.py index 09593f1..c344310 100644 --- a/brozzler/ydl.py +++ b/brozzler/ydl.py @@ -19,7 +19,7 @@ limitations under the License. import logging import yt_dlp from yt_dlp.utils import match_filter_func -import brozzler +import brozzler_ia as brozzler import urllib.request import tempfile import urlcanon diff --git a/setup.py b/setup.py index eea2d5d..2710be0 100644 --- a/setup.py +++ b/setup.py @@ -33,7 +33,7 @@ def find_package_data(package): setuptools.setup( - name="brozzler", + name="brozzler_ia", version="1.6.1", description="Distributed web crawling with browsers", url="https://github.com/internetarchive/brozzler", @@ -41,9 +41,9 @@ setuptools.setup( author_email="nlevitt@archive.org", long_description=open("README.rst", mode="rb").read().decode("UTF-8"), license="Apache License 2.0", - packages=["brozzler", "brozzler.dashboard"], + packages=["brozzler_ia", "brozzler.dashboard"], package_data={ - "brozzler": ["js-templates/*.js*", "behaviors.yaml", "job_schema.yaml"], + "brozzler_ia": ["js-templates/*.js*", "behaviors.yaml", "job_schema.yaml"], "brozzler.dashboard": find_package_data("brozzler.dashboard"), }, entry_points={ diff --git a/tests/test_cli.py b/tests/test_cli.py index 567260f..f399c09 100644 --- a/tests/test_cli.py +++ b/tests/test_cli.py @@ -17,6 +17,7 @@ See the License for the specific language governing permissions and limitations under the License. """ +import brozzler_ia as brozzler import brozzler.cli import pkg_resources import pytest diff --git a/tests/test_cluster.py b/tests/test_cluster.py index 51d78e3..8f54b55 100644 --- a/tests/test_cluster.py +++ b/tests/test_cluster.py @@ -26,7 +26,7 @@ import os import socket import doublethink import time -import brozzler +import brozzler_ia as brozzler import datetime import requests import subprocess diff --git a/tests/test_frontier.py b/tests/test_frontier.py index 760962d..1184e89 100644 --- a/tests/test_frontier.py +++ b/tests/test_frontier.py @@ -26,7 +26,7 @@ import time import doublethink import pytest -import brozzler +import brozzler_ia as brozzler args = argparse.Namespace() args.log_level = logging.INFO diff --git a/tests/test_units.py b/tests/test_units.py index b7a785f..7d5008a 100644 --- a/tests/test_units.py +++ b/tests/test_units.py @@ -21,7 +21,7 @@ import pytest import http.server import threading import os -import brozzler +import brozzler_ia as brozzler import brozzler.chrome import brozzler.ydl import logging